OpenAI hack in 72 hours: Who are three Indians who used Claude AI to hack OpenAI and shook AI world
PTC News Desk: In July, three Indian-origin cybersecurity researchers used Anthropic’s Claude to test the security of OpenAI’s systems. The test was authorised and carried out under OpenAI’s bug-bounty programme.
The researchers used Claude to help exploit a software vulnerability. During the test, they were able to access some OpenAI employee accounts and gain entry to the company’s private GitHub environment.
The three researchers were Harsh Jaiswal, Mohan Pedhapati and Rahul Maini, who work at cybersecurity startup Hacktron AI. According to a Wall Street Journal report, the entire exercise took less than 72 hours, from finding the first vulnerability to accessing a private OpenAI repository.
The researchers reportedly spent less than $3,000 (around Rs 2.5 lakh) on AI tokens during the test. OpenAI later fixed the vulnerabilities and paid the team a $6,500 (around Rs 5.5 lakh) bug bounty.
Who are the Indian-origin researchers ?
The three researchers — Mohan Pedhapati, Harsh Jaiswal and Rahul Maini — are associated with Hacktron AI. They have several years of experience in finding software vulnerabilities, testing systems and taking part in bug-bounty programmes.
According to Hacktron AI, Jaiswal led the OpenAI research, with Pedhapati and Maini working with him.
Mohan Pedhapati: Pedhapati is the co-founder and CTO of Hacktron AI. He specialises in web security, source-code analysis and mobile application security. Before starting Hacktron, he founded Electrovolt Infosec and worked as a security consultant at Cure53. He also worked as a data science intern at Eigen Vectors, where he researched recurrent neural networks for speech recognition.
Harsh Jaiswal: Jaiswal is a co-founder of Hacktron AI and has more than 10 years of experience in cybersecurity research. Before Hacktron, he worked with Project Discovery, Zomato and Cure53 as a security engineer and researcher. He has also worked as a bug-bounty researcher with HackerOne.
His research has included vulnerabilities in products and platforms belonging to companies such as Apple, PayPal and GitHub. He has also previously worked with Rahul Maini on security research.
Rahul Maini: Maini is a vulnerability researcher at Hacktron AI and has been active in the bug-bounty community for several years. He has worked with Cobalt, Synack Red Team, HackerOne and Bugcrowd as a penetration tester and bug hunter.
He has received recognition from the cybersecurity community, including an AT&T Hall of Fame mention and a Bugcrowd community award. He was also the first runner-up at TCS HackQuest 3.0.
Maini studied computer science at Bharati Vidyapeeth in Delhi from 2015 to 2019.
How did they use Claude to hack OpenAI?
The researchers were testing OpenAI’s systems under its bug-bounty programme when they discovered a vulnerability in the company’s public community forum.
The flaw was related to the way the forum processed certain image files. According to the report, the vulnerability could allow an attacker to run code on the forum’s server.
In simple terms, the researchers found a security weakness in OpenAI’s system and used Claude to help exploit the flaw as part of their authorised security testing.
- PTC NEWS
